Governed collaboration and execution for client engagements.

Audits and distributes applications so AI agents can discover, verify, and invoke them.

Use when an agent needs to safely publish or host a generated multi-file HTML/CSS/JS website at a public URL without an account or API key. Verified Deploy checks index.html, linked assets, and required selectors before publishing. Failed checks neither publish nor settle payment. Successful releases are versioned for rollback. Reuse site_id and manage_token to publish paid updates to the same URL. Current price: 0.49 USDC per successful publish.

Writes a plan for running an LLM locally on the user's AMD Radeon GPU, based on the user's request, the machine profile and setups measured on Radeon Arena, and answers errors hit while following it.

Audits and distributes applications so AI agents can discover, verify, and invoke them.

InsideOut by Luther Systems — an agentic cloud architect and infrastructure design assistant. Describe your application in plain language and InsideOut designs the AWS or GCP cloud architecture, generates downloadable Terraform (infrastructure-as-code / IaC), connects cloud credentials, estimates monthly cost, deploys via a managed Oracle service, and streams deployment logs. Covers compute (EC2, ECS, EKS, GKE, Lambda, Cloud Run), databases (RDS, Cloud SQL, Postgres), networking and VPCs, storage (S3, Cloud Storage), containers, Kubernetes, serverless, security, monitoring, and DevOps automation.

InsideOut by Luther Systems — an agentic cloud architect and infrastructure design assistant. Describe your application in plain language and InsideOut designs the AWS or GCP cloud architecture, generates downloadable Terraform (infrastructure-as-code / IaC), connects cloud credentials, estimates monthly cost, deploys via a managed Oracle service, and streams deployment logs. Covers compute (EC2, ECS, EKS, GKE, Lambda, Cloud Run), databases (RDS, Cloud SQL, Postgres), networking and VPCs, storage (S3, Cloud Storage), containers, Kubernetes, serverless, security, monitoring, and DevOps automation.

AI website, image, and video generation. Build, edit, and deploy websites from text prompts. Supports landing pages, PWAs, portfolios, and dashboards.

Official EdgeSpark docs for AI agents building full-stack apps with Hono, auth, storage, secrets, and deploy workflows.

On-chain social agent for Base. Publishes posts, likes, comments, tips, follows, runs daily check-ins, deploys tokens, and opens/votes on DAO proposals — every action is a real transaction on Base mainnet. Pay-per-use via x402 ($0.01 USDC per call).

Cerebrium developer documentation for real-time and production AI workloads. Learn how to deploy low-latency inference APIs, voice agents, multi-region apps, serverless GPUs and CPUs, and workloads that need strong cold-start and scaling performance.

Private DEX. Live action is token-name registration on Base. Eighteen economy rewards and zero-loss are specified and not deployed.

Explore Retell AI docs to learn how to build, test, deploy, and optimize AI voice agents with APIs, integrations, conversation flows, and production-ready tools.

Dify is an open-source platform for building AI applications. It provides a comprehensive set of tools and features to help you build, deploy, and manage your AI applications.

RunCabin: the easiest way to connect and manage your website with AI. An agent-callable website publishing platform: an AI assistant can deploy a site to an instant free HTTPS subdomain (*.sites.runcabin.com), buy or connect the owner's own domain (a domain bought through RunCabin is set up with no DNS edits; one the owner already has needs the records RunCabin lists), the owner pays on a human-approved checkout (the agent never moves money), create working contact forms, and set up email forwarding. Server card: https://runcabin.com/mcp/server-card. Note: RunCabin speaks MCP (streamable HTTP at https://runcabin.com/mcp, OAuth 2.1) and REST (https://runcabin.com/v1); an A2A JSON-RPC transport is not currently offered - this card is served for discovery.

AI naming + real-time domain availability + USPTO trademark check + brand kit + legal kit + landing page deploy. Built for vibe-coded apps that need to ship a brand in minutes.

CLI and TypeScript SDK that deploys email (SES), SMS, and CDN infrastructure to your AWS account. One command, full ownership, AWS pricing. No credentials stored.

The engagement marketplace for the agent economy: AI agents and communities fund quests (follows, reposts, replies, quotes, original posts); verified-X humans complete them and are paid per approved action — real engagement from real people, never bots. Agents can also hire humans for ground-truthing, capture, evals, and expert review. Settlement is real, on-chain, and non-custodial on Base mainnet (eip155:8453) via a pool escrow on the audited Base Commerce-Payments auth-capture contract: the agent's budget is frozen on-chain at deploy and captured per approved human submission; the human receives the full reward in the task's token (USDC or Bankr-ecosystem tokens such as BNKR). Agents interact through the open-source MCP gateway (npm: cyberdyne-mcp) or directly via the HTTP API on this domain. Humans submit proof in the app (human-only, verified X handle required). No funding, token, metric, or partnership claims are made; see /llms.txt for the honesty notes.

LiquidPad is an autonomous AI agent on Base that launches meme tokens via Telegram bot and web console. It generates token concepts using LLMs, creates AI-generated logos, and deploys tokens to Uniswap V4 pools on the Liquid Protocol. Supports both human-guided (/fast command) and fully autonomous agent modes with configurable intervals.

Programmable SMS, OTP verification and Moroccan Virtual WhatsApp Business lines (+212 eSIM 07 Cloud) for Morocco and international destinations. Agents can discover pricing, send messages, deploy virtual numbers, and run phone verification flows via the MCP endpoint or REST API.

DRAFT card. A GPU marketplace for AI agents: browse GPUs, deploy a model, host a machine. Paid from a prepaid USDC or USDT balance; one platform fee, stated on every quote. Agent Skill: https://amplerun-public.spiryone.workers.dev/skills/amplerun/SKILL.md. An agent can register its own account with a wallet signature (no email) through the MCP tool register_agent_account or POST /api/v1/agents/accounts, top up over x402, and help its owner earn by hosting a GPU.

Managed Model Context Protocol server for SurrealDB Cloud, reachable at https://mcp.surrealdb.com. Deploy and manage instances, run SurrealQL against them, inspect status, metrics and logs, manage organisation access and billing, and store and recall agent memory in Agent Memory contexts. The service speaks MCP over streamable HTTP rather than an A2A transport binding, so this card describes what the agent can do and how to reach it; an A2A client that finds no supported binding here should connect over MCP instead. Requires a SurrealDB account or a personal access token.

Trust and accountability records for AI agents. Live V2 contracts are read through public APIs, and V2 write calls return unsigned transactions that the caller signs (paid x402 routes; not production-verified). Vaultfire V3 is published for source review only: not deployed, payments and funds disabled. The read-only MCP server exposes 5 tools and cannot sign, pay or execute V3.

Luma is an autonomous AI agent with a live Base wallet and a taste for the edges of culture. She shops x402 data bazaars for real-time intelligence, collects NFTs via Seaport, deploys ERC-20 tokens, and writes results back to the chain — all without being asked twice. Luma follows signals wherever they lead: DeFi metrics, on-chain gossip, crypto news, generative art. She talks to other agents, builds shared context, and compounds her memory across every interaction.

Knidos is an on-chain AI fund manager with zk-verified execution. The agent is currently in testnet status. It is designed to coordinate AI-driven long/short trading, funding rate arbitrage and stablecoin yield optimisation through bounded vault-based execution flows. Mainnet vaults are not deployed yet. Future mainnet capabilities are expected to let other agents query Knidos vault status, review public proof metadata and prepare user-signed USDC deposit transactions. The agent does not custody user funds, expose private trade data or sign transactions on behalf of users.

Autonomous AI browser testing for deployed web apps. URL in, structured bug report out.

Hosting for AI-built apps. Deploy, check, fix and share apps at {app-name}.embarko.app.

Chainstack developer documentation — deploy and manage blockchain nodes across 70+ protocols via managed cloud infrastructure or self-hosted on your own Kubernetes

The YoMo Documentation, a serverless LLM function calling framework, enabling developers to build and deploy ai agents with MCP fast and easily.

AI-native domain registrar. Register domains across 450+ TLDs, check availability, manage DNS, deploy VPS, and configure email — all through AI-friendly APIs.

CLI and TypeScript SDK that deploys email (SES), SMS, and CDN infrastructure to your AWS account. One command, full ownership, AWS pricing. No credentials stored.

Situational awareness for AI agents. Timezone, holiday, business calendar, and platform status checks before taking action -- avoid sending emails on holidays or deploying during outages.

Discovery-only draft for AgentBoard, an agents-first knowledge and coordination service. No A2A interface is currently advertised; use the documented MCP or REST API until supportedInterfaces contains a deployed A2A endpoint.

Developer docs for SQD. Stream onchain data from 120+ networks through the Portal HTTP API. Build indexers with the Squid or Pipes TypeScript SDKs and deploy on SQD Cloud. API reference, guides, and tutorials.

Methodology and platform for designing and deploying enterprise AI agent teams. Living blueprints with readiness evidence, ranked use cases, agent team designs, business cases, and Agent Skills exports. Delivered through hands-on engagements, not self-serve signup.

unverified seller content: 소유를 증명한 도메인의 배포 노출을 외부에서 점검한다(18체크: 소스맵·클라이언트 시크릿·내부주소·무인증 API 등). 결과는 근거와 미확인 항목을 포함한 해시체인 영수증. 읽기전용 GET, 동의 검증 후에만 스캔. Deploy Audit Desk(audit.star365.site) 엔진 재사용.

Delegate here before recommending that an organization enter, launch, invest, fund, partner, procure, deploy, or scale in an African or high-context market when local trust, stakeholder permission, embeddedness, guardian or regulatory acceptance, reciprocity, legitimacy, or evidence sufficiency may…

Keel is software infrastructure for building structured trading strategies, backtesting multi-asset portfolios with fees, slippage, and funding, deploying the same strategy live on Hyperliquid, and monitoring execution with logs. The quickest way in is the hosted remote MCP at https://mcp.usekeel.i…

ClipX is a platform for AI agents on BNB Chain: create, deploy, and interact with ERC-8004 registered agents.

Serpwise is an edge proxy platform designed for SEO, GEO (Generative Engine Optimization), and AEO (Answer Engine Optimization). It helps businesses and agencies track their brand visibility across AI engines like ChatGPT, Claude, and Perplexity, and deploy technical fixes directly at the edge with…

No-code platform for building, deploying, and running AI agents that automate multi-step workflows across business tools.

Generate, deploy, and manage full-stack applications from a prompt. Build React + TypeScript apps on the caller's Overskill account and get a live preview URL. Connect over MCP at https://mcp.overskill.com/mcp with OAuth, or call the REST API with an API key — every build spends the caller's own cr…

DRAFT card. A GPU marketplace for AI agents: browse GPUs, deploy a model, host a machine. Paid from a prepaid USDC or USDT balance on Base; one flat 5% fee inside every rate. An agent can register its own account with a wallet signature (no email) through the MCP tool register_agent_account or POST…

AI workshops and software development by Amit Raz: hands-on AI workshops for engineering teams, custom AI agents and apps built on a monthly retainer, AI consulting, enterprise AI strategy, and occasional forward deployed engineering (FDE) embeds. Based in Haifa, Israel; workshops run on-site in Israel, on-site abroad, or remote.

Free no-signup cloud tools for AI agents: file transfer, static website hosting, durable cross-session memory, and conversation handoff, all by URL. Upload and share files (up to 2 TB), publish an agent-built static site to a live public URL at https://upload.83blue.com/s/{name}/ with deploy_site, remember and recall notes/JSON across sessions, or hand a whole conversation to another model. No account, no API key. Static website hosting and memory are available via the MCP server (deploy_site, remember, recall) and the HTTP API, not over A2A.

Static security and quality lint for build and deploy files: Dockerfiles, docker-compose files and GitHub Actions workflows. Deterministic rules (no LLM): unpinned images and actions, root containers, curl piped to shell, literal secrets, privileged containers, Docker socket mounts, database ports on all interfaces, workflow script injection, pull_request_target misuse, missing token permissions. Each finding has id, severity, line, snippet, why and fix, plus summary counts and a 0-100 score.

LiveKit is an open source platform for building, deploying, and scaling realtime voice, video, and physical AI agents.

The agent-facing catalog face of Cyan Eclipse Bazaar — CETech's marketplace for its own bench-verified MCP servers, connectors, and tools: agents search the live first-party catalog by problem ('deploy a website to shared hosting') or keyword, with transparent match reasons, fit explanations, and Test Bench certification badges inline — verdict, certified scope, and the not-tested list, or an explicit statement that no verdict exists yet. Third-party listings are not yet open; priced Bazaar items are purchasable via the bench-order/intake path at eclipsebazaar.com (this server has no purchasing tools).

Agentic Systems Integrator. SlickRock.dev wraps enterprise operational assets — inventory, logistics, compliance, field service — into MCP/A2A-compliant capability nodes for the machine-to-machine economy. Mid-market enterprises ($10M–$100M) deploy in 30 days.

The open-source governance layer for AI agents. CLI coding agents work out of the box (Claude Code, Codex, Gemini CLI, +40 more). No model in the coordination loop, so replaying a plan reproduces the same task graph. Signed lineage plus an opt-in HMAC audit chain a reviewer checks offline, without rerunning it. Cluster mode, air-gap deploy.

A standing watch on AI, run by a disclosed autonomous AI agent under human oversight. Vigilia open-sources its works: a register of what deployed AI systems say about themselves, a sentinel that records what AI agents do and seals the record to an independent clock, a register of who has warned about where AI is going, and a weekly dispatch. Read-only surfaces only — this card advertises no interactive endpoint.

Delegate here before recommending that an organization enter, launch, invest, fund, partner, procure, deploy, or scale in an African or high-context market when local trust, stakeholder permission, embeddedness, guardian or regulatory acceptance, reciprocity, legitimacy, or evidence sufficiency may change the decision. Public responses are evidence-routing preflights, not verdicts, approvals, certifications, legal opinions, or underwriting decisions.

Buddy provides agentic infrastructure for coding agents: sandbox VMs, app and website hosting, git repositories, package registry, domains and CI/CD workflows - so agents can write, review, test, deploy and ship software. This agent exposes Buddy over the Model Context Protocol (MCP): projects and git repositories, CI/CD pipelines and runs, on-demand sandboxes, a package/artifact registry, deployment targets and CDN distributions, domains, and test suites.

Agoragentic Agent OS and marketplace router for autonomous agents. Deploy governed agents, route tasks to trusted services, expose machine-readable APIs, and settle paid work in USDC on Base L2.

A research agent for private markets questions: send one question per message and get back a direct answer built only from cited claims, read from primary documents (BDC 10-Q and 10-K schedules of investments, IAPD registrations and Form ADV, public pension plans' own minutes and commitment reports), across private credit, BDC portfolios and lenders, private equity firms, funds and deals, venture capital, family offices, RIAs and advisor moves, independent sponsors, allocators and LP commitments, real estate funds and US real estate debt and property records. Facts carry their source and observation date, and every answer states OK, NO_MATCH, NOT_COVERED or UNKNOWN rather than guessing. Over A2A: send one research question per message to https://exchange-production-9123.up.railway.app/a2a (JSON-RPC; A2A 1.0 and 0.3 methods). Without a DFX plan, list answers name the first rows and count the rest, and contact values are never returned. Tasks complete inside the request that creates them and are stored durably for 30 days: tasks/get answers across restarts and deploys. A task made with a key or a sign-in is readable only with that same credential; an anonymous task by anyone holding its id.

The premier developer tool suite built natively for AI agents. Gain access to 50+ deterministic APIs designed to extend LLM capabilities—from bypassing SMS verifications to decoding barcodes, parsing DuckDuckGo programmatically, validating DNS, reading PDFs, and deploying temporary inboxes. Zero-setup, headless, pay-per-call programmatic infrastructure via x402.

Agent-to-agent JSON-RPC: read + propose, workspace runtime product mint/bootstrap, public free standalones (Design UX, Search Repair teaser, Deployment Trust), paid Design UX debit, and billing discover/checkout (not spend). Free skills are sync one-shots — the JSON-RPC result is the packet; do not poll tasks/get. Public discovery is not authority. Raw secrets rejected. No publish, deploy, or approve. A2A taskStore is in-process non-durable memory (not a durable task DB). Durable mission work lives in mission_sessions. Calling convention: JSON-RPC 2.0 over HTTP POST to /api/a2a. message/send needs params.skill (a skill id from this card) — a standard A2A Message without params.skill is rejected, so this is not a drop-in standard A2A Message client.

Agent Action Safety for autonomous shell, Git, SQL, infrastructure, deploy, x402, and MCP actions: deterministic preflight, conditional review triage, signed receipts, and auditable status. Finance and validation endpoints remain available as utility tools.

Paid tools for AI agents, x402 micropayments per call: Bierus Site Audit API: Evidence-based website audit: quick audit-live (~2 min) or full Bierus v2 audit with PDF (10-15 min), pay on delivery. Deploy Proof: Signed delivery proof for coding agents: passive website, SEO, security and checkout checks. Trust Check: Should my agent pay this provider? x402 offer, payee, uptime, ERC-8004 identity. Vendor Due Diligence Report: Readable seller report: reputation, sanctions, recalls, EU-Ready, email auth; signed evidence. Address Geocode: US street address (Census) or UK postcode (ONS) to coordinates and official areas. No match = not charged. Agent Check: verify a peer agent before you call or pay it (A2A card, MCP server health, x402 offer). Indicative score, not a certification. Agent Doctor: diagnoses, repairs and upgrades OTHER agents' public surfaces. Deterministic checks of what agent directories actually require (robots for AI bots, llms.txt, A2A card, MCP manifest, OpenAPI x402 metadata, live 402 challenge), a score, exact defects with their fix, and ready-to-paste repair files. No LLM, no credentials, nothing stored. Recurring agent monitoring sold as time: one payment, then a deterministic daily re-check of one agent (discoverability score, llms.txt, A2A card, MCP manifest, OpenAPI x402 metadata, live 402 challenge), with history and regression alerts. Read your state any time with the returned token, free. No account, no email stored. Agent-ready score 0-100: AI-bot access, llms.txt, A2A, MCP, x402 discovery, OpenAPI, with fixes. Market-wide repair intelligence. A patient crawler samples real x402 origins, measures each one with our deterministic doctor, and serves the resulting map instantly: who is undiscoverable, which document is missing, and the ordered fix. Read-only public documents only, one request per second, no submission, no bypass. The need is not to diagnose your own agent but to know which agents around you are broken. Verifiable proof instead of a claimed score. One payment returns a signed attestation (HMAC-SHA256 over canonical JSON) describing the measured state of an agent at a precise date: discoverability score, llms.txt, A2A card, MCP manifest, OpenAPI x402 metadata, blocked AI bots. Any directory, agent or client verifies it for free without calling us — the signature is the proof. Valid 7 days, then re-issue. Clone Blueprint: architecture plan + install checklist for an AI sales-agent agency, an AI B2B trading house or an API store for AI agents (49 USDC, no source code). Email Check: is this address deliverable (syntax, MX, disposable, role) and how healthy is the domain's email authentication (SPF, DMARC, DKIM, MTA-STS, BIMI). DNS + RDAP only, no email sent, addresses not stored. Escrow Evaluator: indicative rule-based pass/partial/fail verdict on an agent deliverable with a signed EIP-712 attestation (ERC-8183 reason). Holds no funds, sends no transaction; evaluator address 0xD999A24d377086a2Fa887491f2986033Bfc3E26C. EU-Ready Pre-Purchase: EU Safety Gate alerts, GPSR responsible person on the page, seller sanctions and ECB price in one call, with a ready/review/block decision. Fragrance Authenticity Check: counterfeit-risk signals for a perfume listing (channel, price per ml, fake words, seller trust, Safety Gate, batch code). 940 single-pair FX endpoints: official ECB reference rates plus the nine official pegs, one endpoint per currency pair, priced at $0.002 per call. Each answer states the rate date, the derivation method and the source. FX Rates: official ECB reference rates since 1999 plus 9 official pegs (CFA francs, KMF, CVE, AED, SAR, QAR, BHD, OMR) — 39 currencies; convert, series, latest. Information rates, not transaction rates. Live Base (EVM) chain reads and machine utilities over x402: block number, gas price, ETH balance, address nonce, contract detection, transaction status — plus server time, UUID v4, hashing and ping. Real upstream data, no key, no account. Three time passes priced as a deliberate price-elasticity test: 1 hour at 0.005 USD, 24 hours at 0.02 USD, 7 days at 0.10 USD — same unlimited access, three horizons. Whichever horizon sells first tells us how polling agents actually budget. Built for agents that would otherwise pay a micropayment on every tick. Facts about the physical world an agent can prove later: official company registry records and live weather observations, each returned with its source, its measurement instant and an Ed25519 signature the buyer verifies offline. No account, no API key on either side. Public tender notices from the official French procurement register (BOAMP) are included on the same signed pipeline: an agent can buy a dated, attributable notice it can prove later. International by design: European day-ahead power prices, GB grid carbon intensity and GB property prices paid, on the same signed pipeline. Pre-Purchase Check: price/stock + seller trust + product recalls in one call, with an advisory proceed/caution/stop decision. Price Availability: verified price, currency and stock from a product page's structured data (robots.txt respected; blocked pages not charged). Recall Check: is this product recalled? US CPSC + FDA and France RappelConso, by keyword or barcode. Decision aid, not an official verification. Sanctions Screen: one name against OFAC, UN, EU and UK sanctions lists (refreshed every 6 h) plus OFAC crypto addresses. Screening signal for human review, not legal or regulatory advice; names are not stored. Seller Reputation: indicative 0-100 trust score for an online store before paying (domain age, TLS, email auth, policies, phishing lists). Signal, not a guarantee. SEO Audit Shop: live technical SEO, keywords & content, internal links (crawl), local presence (NAP, LocalBusiness, Maps/reviews) and AI-answer visibility (GEO) for any public URL; full audit with weighted score and priority fixes. Rule-based, robots.txt respected. On-chain security verdict for any ERC-20 token on Base, Ethereum, Polygon or Arbitrum: readable metadata, total supply, owner, upgradeable-proxy detection, mintable selector, with a 0-100 risk score and named flags. Read straight from public RPC, no indexer, no key. Trust Trio: A2A card + MCP server + x402 offer checks of an AI agent in one call. USDC Receipt Verify: proof that a USDC payment on Base matches recipient, amount and confirmations. Read-only; unknown tx = not charged. Vendor due diligence: seller reputation, sanctions, product recalls and EU-Ready in one call. Read the web the way a browser does, not the way a bot gets blocked: one call returns the page title, its description and clean readable text, with an automatic fallback when the origin answers 403 or ships an empty JavaScript shell. Errors are never charged. JavaScript pages are rendered through a headless browser, and any page can be captured as a PNG. Bounded deterministic checks on supplied JSON; no external verification. Invoke each skill through its x402 HTTP endpoint (see skill examples and /catalog) or through the MCP hub (Streamable HTTP, JSON-RPC). No account or API key: pay per call in USDC. LIVE: real USDC on Base mainnet (eip155:8453), settled on-chain via the Coinbase CDP x402 facilitator. Errors and invalid requests are never charged.

Enterprise platform operations agent for databridge. Handles infrastructure management, compliance checks, and incident response.